Simple Analysis of Police Policies in the U.S.
By Trisha Sanghal, Coco Sun, and Sina Ghandian
Background
Over the past few weeks, our team has been researching various police use of force policies and analyzing their relationships with crime rates in American cities. We wanted to build on existing research into how these policies relate to the number of deaths caused by police departments, such as the findings in this document by Campaign Zero. We explored whether or not certain policies are correlated with certain levels of crime. Are cities with greater crime rates more or less likely to implement restrictive use of force policies? Which policies seem to make the biggest impact on crime rate, or vice versa? These are some of the questions we hoped to answer.
To accomplish this, we chose eight common policies and eighty-eight cities of varying size and geographic location to analyze. (This includes many major cities, such as New York and Los Angeles.) The policies we looked into were:
- Bans chokeholds and strangleholds: In cities where this policy is implemented, the use of chokeholds or strangleholds is prohibited except as an alternative in situations where lethal force is justified. (Approximately 31.8% of the cities we analyzed enforced this policy.)
- Duty to intervene: Officers are required to intervene if they see other officers use excessive or unnecessary force. (Approximately 54.5% of the cities we analyzed enforced this policy.)
- Has use of force continuum: Officers must respond with a level of force that is appropriate for the situation, moving quickly from one part of the continuum to another if need be. (Approximately 83.0% of the cities we analyzed enforced this policy.)
- Requires comprehensive reporting: Here, officers must document any use of force promptly, completely, and accurately in an appropriate report. This policy also requires officers to report any excessive force they witness. (Approximately 29.5% of the cities we analyzed enforced this policy.)
- Requires de-escalation: When feasible, as an alternative to force, officers should utilize de-escalation and crisis intervention tactics. (Approximately 48.9% of the cities we analyzed enforced this policy.)
- Requires exhaust all other means before shooting: After all other alternative means have been exhausted, officers can consider the use of firearms only in defense of life or to prevent serious bodily harm to the officer or others. (Approximately 42.0% of the cities we analyzed enforced this policy.)
- Requires warning before shooting: Prior to the use of force, an officer should identify themselves as a peace officer and warn that deadly force may be used. (Approximately 69.3% of the cities we analyzed enforced this policy.)
- Restricts shooting at moving vehicles: Shooting at any part of a moving vehicle in an attempt to disable it is restricted. An officer is only allowed to shoot at a moving vehicle if they believe there are no other means available to avert the threat, or if deadly force is directed at the officer or others. (Approximately 18.2% of the cities we analyzed enforced this policy.)
Findings
Since we were primarily interested in the correlation between crime rate and policy implementation, we plotted the normalized violent crime rate in each included city against the policy of interest. (We used normalized crime rates as opposed to using totals because city populations can also influence the number of crimes within the city, and normalization allows for more consistent comparisons.) In the plots below:
- Each circle represents a city’s location.
- The size of the circle represents the normalized crime rate in that city, or the number of violent crimes per 10,000 people.
- The color of the circle represents whether or not the highlighted policy has been implemented in that city, where yellow indicates that it has and blue indicates that it hasn’t. (The intensity of the color in the second image represents the number of restrictive policies enforced by that city’s police department.)
From the above visualizations, it may seem as though cities with higher violent crime rates enforce more restrictive use of force policies. This would make intuitive sense since we might expect officers in cities with high crime rates to face more situations in which they must assess their use of force (and therefore might expect to see more restrictive policies in place to govern this use of force). However, after further digging, our analysis revealed a low correlation coefficient of approximately 0.102. This indicates that there is a weak, positive linear relationship between the number of violent crimes in a city and the number of restrictive use of force policies implemented there. We also found that about 60.2% of the cities analyzed have only two to four of the eight restrictive policies implemented, and only 17.0% of the cities have six or more policies implemented. In other words, violent crime doesn’t explain restrictive policy, and not many cities have these policies implemented. With this in mind, it’s important to note that police departments are complicated entities. There are several other factors that may influence the relationship between crime rates and use of force policies. This includes, but isn’t limited to:
- Basic recruit training and in-service training: A department’s procedures relating to how and how often training should be conducted (e.g. training related to use of force policies, de-escalation, and threat perception).
- Investigations of deadly force incidents: A department’s procedures for how investigations of deadly force incidents should be handled and the scope of these investigations (e.g. whether investigations should place more emphasis on a given officer’s tactic or decision making process).
- The relationship between police departments and the local community: How transparent a department is regarding any pending investigations or internal review.
- Racial composition: The proportion of the population in each racial group.
- Economic factors: The distribution of income in the city, and the funding that police departments receive.
Conclusion
From past studies, we have learned that cities with more restrictive use of force policies tend to have fewer cases of excessive violence or death at the hands of police. Our findings have now shown that the number of restrictive use of force policies in place is not highly influenced by the levels of violent crime within a city. In the future, it would also be interesting to analyze how other factors like police budgets come into play when considering the enforcement of these policies and their relationships with crime rates and police violence. There is still a lot to learn and uncover, and we encourage readers to continue to look into this topic!
Sources
- John Miller’s Police Policies Data
- Census Bureau’s City Population Data
- EPD’s Use of Force Policy Explanations
- NIJ’s Use of Force Continuum Explanation
- Philadelphia PD’s Assessment of Deadly Force
